timhigins

5 ms·

31.
by timhigins 1y ago
Note that Trump's Big Beautiful Bill as it passed the House of Reps would bring back 100% expensing of R&D expenses including software development costs/salaries.
32.
by timhigins 1y ago
> Exempted Code > Not all source code, documentation and metadata used in the development of Direct File is included in this repository. Specifically, any code or data that is considered Personally Identifiable Information (PII), Fede
33.
by timhigins 1y ago
The code that defines how the fact graph works is here: https://github.com/IRS-Public/direct-file/tree/main/direct-f... but the actual tax definitions that deal with facts and derived calculation are her
34.
by timhigins 1y ago
Thanks for these comments! Seems their measurement of Blender is off and we need some more in-depth benchmarks.
35.
by timhigins 1y ago
This has been researched quite a lot with the adjacent task of structure from motion (2d videos) - see NeRFs, Gaussian Splatting - and now there are a few models that do 3d from single 2d images fairly well.
36.
by timhigins 1y ago
The coolest thing here might be the speed: for a given scene RenderFormer takes 0.0760 seconds while Blender Cycles takes 3.97 seconds (or 12.05 secs at a higher setting), while retaining a 0.9526 Structural Similarity Index Measure (0-1 wh
37.
by timhigins 1y ago
This looks super cool! Seems like a stronger statistics-based optimization strategy than https://docs.auto-rag.com/optimization/optimization.html . I've got a few questions: Could I ignore all flow runs that have a
38.
Curtis Yarvin wants to replace American democracy with a monarchy led by a 'CEO'(cnn.com)
12 points by timhigins 1y ago|12 comments
39.
by timhigins 1y ago
This paper seems to focus on highly algorithmic/puzzle-like problems, which are not the typical application domain of LLMs, using a <500M parameter model. So my hunch is "reasoning" works much better for math, coding, fact
40.
by timhigins 1y ago
> LLM could hallucinate The job of any context retrieval system is to retrieve the relevant info for the task so the LLM doesn't hallucinate. Maybe build a benchmark based on less-known external libraries with test cases that can
41.
by timhigins 1y ago
See also https://storage.googleapis.com/transparencyreport/report-dow... https://developers.google.com/search/help/about-search-data-... https://docs.google.com/forms/d&#
42.
Google's European Search Dataset Licensing Program, GDPR, and DMA Article 6(11)(truthonthemarket.com)
2 points by timhigins 1y ago|1 comments
43.
by timhigins 1y ago
looks great! but your linked homepage doesn't work: https://colibri-hq.github.io/colibri/
44.
by timhigins 1y ago
Screen studio does make the best-looking demo videos I've seen. Any favorite alternatives? Points for free or open source.
45.
by timhigins 1y ago
One guy simulated Australia's electrical grid for 3 years to show that the entire grid could run off of >98% renewable energy using the installed renewables with just a small amount of storage (e.g. 5 hours of average demand, ~10% o
46.
by timhigins 2y ago
Can you just compute the travel times from each bus/train/tram stop and publish the individual or merged travel times? I guess the merging would be the hard part
47.
by timhigins 2y ago
Do you think this indexing architecture would bring benefits to general web research? If implemented like: planner, searches, index webpages in chunks, search in index, response Would you ever extend your app to search the web or specialize
48.
by timhigins 2y ago
Removing these records from the public internet could likely be considered illegal under the OPEN Government Data Act, 44 U.S.C. § 3506: > (d) With respect to information dissemination, each agency shall— (3) provide adequate notice when
49.
by timhigins 2y ago
Vaccinated: One Man's Quest to Defeat the World's Deadliest Diseases Describes how Maurice Hilleman invented 40 vaccines, including for eight of the most common diseases in the US, over a 36 year career at the Walter Reed Army Med
50.
by timhigins 2y ago
If you have 200 YAML files for a single service and 46 clusters I think you're using k8s wrong. And 5 + 3 different monitoring and logging tools could be a symptom of chaos in the organization. k8s, and the go runtime and network sta
51.
by timhigins 2y ago
Update: While SmolLM was SOTA at the time of release in July, SmolLM 2 1.7B (which is the newest release) is not currently the best model under 2B params on https://huggingface.co/spaces/open-llm-leaderboard/open_l
52.
by timhigins 2y ago
Might be worth updating the title to "SmolLM: state-of-the-art small language model trained on open datasets" (See the first table of https://huggingface.co/blog/smollm for benchmarks) It was fascinating digg
53.
by timhigins 2y ago
This reads like a sponsored/promoted article. Would be nice to see a disclosure if it is or they get affiliate link revenue
54.
by timhigins 3y ago
Any reason for using the term "generative", which may confuse readers and imply generative AI/LLMs? It's a traditional tabular autoML system, though it does learn pipelines from a corpus of Kaggle solutions, and generate
55.
by timhigins 3y ago
Site is down for me, maybe the HN hug of death? Anyone else seeing this?
56.
by timhigins 3y ago
According to Intel's own press release, "Argonne is spearheading an international collaboration to advance the project, including Intel; HPE; Department of Energy laboratories; U.S. and international universities; nonprofits; and
57.
by timhigins 4y ago
What do you mean by this, and what would be their reason for doing so? I've tested a few prompts for scraping and there have been no problems.
58.
by timhigins 4y ago
I find this a good overview of the topic: https://damueller.com/#/blog-post/NNLLs
59.
by timhigins 4y ago
Can you publish some of the prompts that you used?
60.
Named tensors with first-class dimensions for PyTorch(github.com)
4 points by timhigins 4y ago|0 comments

More ›